The mouse has become an invaluable model organism for biomedical research in the post-genome era. The decades of research to uncover linkages between the genome and phenome using mouse models has led to an accumulation of a body of knowledge on the genetic diversity and gene-environment interactions that is yet to be matched for any other mammal, including humans. Importantly, the mouse has been the stalwart of mechanistic research in toxicology, including studies of liver toxicity by ethanol. Here, the investigators are proposing to use metabolomics and mouse models in a new systems biology approach for molecular dissection and discovery of biological pathways underlying the susceptibility to ethanol-induced liver injury. The investigators will test the hypothesis that by combining a state-of-the-art in vivo model of liver toxicity and the investigators' prior knowledge of the mechanisms of ethanol-induced liver injury, novel metabolic, genomic and pathologic analyses with genetic diversity of mouse inbred strains will define a "liver toxicity susceptibility state." A set of inbred strains that provides a strong link to the Mouse Phenome Project will be used in these studies. The first aim will be to develop and test computational strategies for integration of metabolomic data with other multi-dimensional data types for studies into the mechanisms of liver injury. The optimal approaches will be applied in the second aim to a panel of fourteen inbred mouse strains to produce mouse strain-specific baseline metabolic and gene expression matrices in the target tissues. The third aim is to produce a mouse strain-specific multi-dimensional toxicological, metabolic and expression data set related to acute and sub-chronic liver injury due to ethanol. Results from this aim will be compared to known gene-environment interactions in these inbred strains to other liver toxicants. Finally, the investigators will perform phenotype-specific genotypic anchoring of ethanol-induced metabolic and expression changes with liver injury. The investigators will develop methods for translating the resulting information into human studies. Data collected in these studies will be deposited into publicly-available databases and mined for unique signatures indicative of common responses to ethanol, other liver toxicants and for specific genetic background dependent responses. The investigators believe that the data and computational strategies produced in this project will be useful far beyond the immediate scope of this research project since it can (i) be integrated and analyzed in conjunction with ongoing projects from the toxicogenomics consortium to identify common species-dependent and -independent responses to a variety of environmental toxins; (ii) provide valuable insights into genetically determined individual variability in response to toxicants; and (iii) identify physiologically relevant animal models for studies on specific patterns of toxicant responses observed in genetically polymorphic human individuals.